If an agent goes down and generates an error on the server, (such as "Get value from agent failed: cannot connect to [[10.X.X.X]:10050]: [4] Interrupted system call") that error is not reset and still appears in the dashboard and is reported via the API if the agent is re-IPed before coming back up again.

This happens relatively frequently in our environment, as we have immutable servers monitored via Zabbix where the hostname is recycled, but the IP address is not. We're monitoring via a Zabbix proxy, it's not clear if the problem is as a result of the proxy or the server.